Dr. Kaeochinda's Stats Portal
Just getting started in PF744 – Statistics I? You may want to follow my video below and install SPSS.
At the proposal level, you will have to determine the sample size (n) of your study by conducting a power analysis. You will need to know your variables (IV and DV) as well as all grouping. If you have your standardize measurements, it would be best to finalize those before conducting a power analysis. You should also have a well-formed research question and hypothesis/hypotheses before attempting the power analysis. When you’re ready and have all a basic idea of the structure of your study, use GPower to conduct your power analysis.
GPOWER for Sample Size Estimation
- GPower: https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower (scroll down to Downloads)
- Minimum alpha level should be set to .05
- Minimum power should be set to .80
- The rest will depend on your hypothesis, groups/grouping, measurements, and expected analysis type
Example APA style write-up for GPower in your methods section: “An a priori power analysis was conducted using GPower version 18.104.22.168 (Faul et al., 2007) to determine the minimum sample size required to test the study hypothesis. Results from the GPower analysis indicated the required sample size to achieve 80% power for detecting a small effect, at a significance criterion of α = .05, was N = 100 for multiple, linear regression.”
Link to Faul et al. (2007) article for your citation: https://link.springer.com/article/10.3758/BF03193146
Of course, more participants is always better!
Basic Statistics Examples
I will be using the following SPSS dataset for the examples below: SPSS SAV File
(Please note that this data is fabricated and no real participants are depicted)
You can also see the table of content at YouTube to skip to various parts of the video above: https://youtu.be/R-8F7fwB4NI
Handy statistics tables of critical values.
What statistics test should I use for my dissertation?
In psychology research, the choice of statistical analysis depends on the type of variables being examined. Generally, categorical variables are analyzed using non-parametric statistical tests, while continuous variables are analyzed using parametric tests. Here are some common combinations of variables and the appropriate statistical tests:
Categorical variables only:
- Chi-square test: This test is used to analyze the relationship between two categorical variables. You would also include contingency tables with observed and expected values.
- Cross-tabs: Not really an analysis but an organization of your data in relationship to each other with frequency or count in each cell.
Continuous variables only:
- Correlation analysis: This test is used to analyze the relationship between two continuous variables. Typically, in psychology, this is only used as a starting point to observe relationship between all variables (although plenty of studies publish based on correlation alone). We typically use this with data that has already been established (sometimes without random selection or random sampling).
- t-test: This test is used to compare the means of two groups on a continuous variable.
Combination of categorical and continuous variables:
- ANOVA (Analysis of Variance): This test is used to compare the means of three or more groups on a continuous variable when the independent variable is categorical.
- MANOVA (Multivariate Analysis of Variance): This test is used to compare the means of three or more groups on multiple continuous variables when the independent variable is categorical.
- ANCOVA (Analysis of Covariance): This test is used to compare the means of two or more groups on a continuous variable while controlling for the effects of one or more continuous covariates.
It’s important to note that these are general guidelines, and the choice of statistical analysis ultimately depends on the specific research question, data type, and assumptions of the statistical tests. It’s always a good idea to consult with a statistician or use statistical software to help determine the appropriate analysis for your research.
A special note on regression analysis. You should not pick regression just because you want to answer a simple research question or hypothesis. Please consider the following.
Regression analysis is a statistical technique used to examine the relationship between two or more variables. In general, both the dependent and independent variables should be continuous variables, meaning that they can take on a range of values and not just a limited set of categories. It is typically used when the researcher wants to understand how changes in one variable affect changes in another variable.
Regression analysis is particularly useful when:
There is a hypothesized or theoretical relationship between two or more variables: Regression analysis can help to confirm or reject hypotheses about the relationship between variables, and to estimate the strength and direction of the relationship.
There is a need to predict one variable based on another variable: Regression analysis can be used to develop predictive models that can be used to forecast future values of one variable based on the values of another variable. For example, the sales of an album based on the number of times the song is played on the radio or advertised on the radio.
There are multiple predictor variables: Multiple regression analysis can be used to examine the relationship between one dependent variable and multiple independent variables. This can help to identify the strongest predictors of the dependent variable.
There is a need to control for the effects of other variables: Regression analysis can be used to control for the effects of other variables that may be related to the dependent variable, allowing for a more accurate estimation of the relationship between the dependent variable and the independent variable.
There is a need to estimate the magnitude of the relationship between variables: Regression analysis can be used to estimate the strength of the relationship between variables, as well as to determine the direction of the relationship (i.e., positive or negative).
Regression analysis is a useful statistical technique for exploring the relationship between variables, making predictions, and estimating the strength of the relationship between variables. It can be particularly useful when there are multiple predictors, a need to control for the effects of other variables, or a need to estimate the magnitude of the relationship between variables.
In the case of a simple linear regression, where there is only one independent variable, a continuous independent variable and a continuous dependent variable are ideal. For example, you might use a simple linear regression to examine the relationship between a person’s height (the independent variable) and their weight (the dependent variable). Both height and weight are continuous variables that can take on a range of values, and a simple linear regression can be used to estimate the relationship between the two variables.
In multiple regression, where there are two or more independent variables, all of the variables should be continuous. For example, you might use multiple regression to examine the relationship between a person’s weight (the dependent variable) and their height, age, and exercise frequency (the independent variables). All of these variables are continuous, and multiple regression can be used to estimate the relationship between the dependent variable and the multiple independent variables.
It’s important to note that some types of regression analysis, such as logistic regression or Poisson regression, can be used with categorical or count data as the dependent variable, respectively. However, for the most common type of regression analysis, which is linear regression, continuous variables are generally required.